While there has been a recent upsurge of research in other anxiety disorders, social phobia has been neglected. The few pharmacological trials involving social phobics are uninterpretable because of small N's, inadequate dosages, lack of uniform or specified diagnostic criteria, and most important, a tendency to study mixed phobic populations and not seprately report social phobic outcome. Also, epidemological, familial and biological challenge studies of social phobia are entirely lacking. The existing literature and our preliminary data both suggest the following: Social phobia seriously impairs work and social functioning, and is associated with several secondary psychiatric syndromes, especially alcohol abuse. Social phobia may be dramatically responsive to MAO inhibitors and beta adrenergic blockers. Demographic, clinical and psychobiological data suggest patients who meet DSM-III criteria for social phobia (primary social phobics) are distinct from panic disorder or agoraphobic patients, many of whom develop secondary social phobic features. Within primary social phobia, patients with specific limited social fears (i.e. public speaking) may differ from those with more generalized pervasive social anxiety. We propose a placebo controlled evaluation of the MAOI phenelzine and the beta bocker atenolol in 90 DSM-III social phobics, to be carried out over 4\years. Eighty panic disorder or agoraphobic patients with secondary social phobic features will be studied as a comparison group. Responders and partial responders to acute treatment will enter a maintenance phase. Responders to phenelzine or atenolol in the maintenance phase will enter a placebo controlled discontinuation phase. Patients will also undergo lactate and other biological challenges and family study as part of already funded studies. Various patient subtypes will be compared in terms of treatment response and other parameters. The clinical significance of this program is to test the efficacy of two promising treatments for various subtypes of social phobia. The theoretical significance lies in the fact that controlled data will be provided on the contribution of clinical subtype, presence or absence of panic attacks and initial level of depression to response of socially anxious patients to MAOIs and beta blockers.